On Fri, 17 Dec 2004, Peter Stuge wrote:
>On Fri, Dec 17, 2004 at 08:30:41AM +0100, Andreas Aardal Hanssen wrote:
>> Now if someone can come up with a database that can store random data
>> of sizes varying from 512 bytes to several megabytes, at 10-15
>> arrivals a second, I'd be willing to change my opinion. ;)
>It would definately require creative "routing" based at least on
>size, using different types of stores for different size mails.
>But with the numbers you quote there's simply no easy solution.
>Sure, you can throw hardware at it and try to parallellize, but that
>only works a certain distance too.

Sure, but at least it's a linear scale, and byte-by-byte, the performance
of direct disk access is a magnitude better without the database :-).

Huge loaded systems are great fun, hehe.

Andy

--
Andreas Aardal Hanssen   | http://www.andreas.hanssen.name/gpg
Author of Binc IMAP      |  "It is better not to do something
http://www.bincimap.org/ |        than to do it poorly."

Reply via email to